Imagens das páginas



644. Weather is the condition of the air at any time as regards heat, moisture, wind, rain, cloud, and electricity; and a change of weather implies a change in one or more of these atmospheric elements. From the direct bearing which weather-changes have on human interests and pursuits, ther have been closely observed from the earliest times, in order that their approach might be predicted with some degree of confidence. The strong craving in the public mind for this knowledge is attested by the prognostics current in every language, which, amid much that is shrewd and of practical value, embody more that is vague, and not a little that is absurd. Any reference to Moore and other almanacmakers is unnecessary, except as testimonies to a widespread ignorance of even the most palpable elements of physical law, which is a disgrace to the educational system of the country. When prognosticators of higher pretensions appear before the public with revelations, weeks or months beforehand, of fine or stormy weather fraught with great advantage or incalculable disaster to agricultural and other interests, it is curious to note how their predictions are laid hold of by the newspapers and scattered broadcast over the country.

645. The truth is, no prediction of the weather can be made, at least in the British Islands, for more than three, or perhaps only two days beforehand; and any attempt at a longer prediction is illusory. But though no prediction of

the weather weeks or months beforehand can be made with any pretension to trustworthiness, yet guesses or surmises may be formed which are not without value. All prediction based on solar or other astronomical causes, if not misleading, is useless. Investigations appear at present to point to a connection between the positions of the planets on the one hand, and the sun's spots, terrestrial magnetism, and the aurora, on the other hand. Nothing, however, could be inferred from such a connection, even were it conclusively established, that could be turned to account in predicting the weather likely to occur in a particular country within a specified time. The only safe guides we can have in attempting to forecast the weather for some time are averages based on terrestrial observations.

646. Of this class may be mentioned the interruptions which occur in the regular march of temperature in the course of the year, of which some account has been already given at page 140. Thus, since in Scotland at least, cold weather prevails some time in the second week of February, April, May, August, and November, and in the end of June; and warm weather in the second week of July and August, and in the beginning of December, it follows that, when at these times the weather begins to grow cold or warm, a continuance of such weather may be expected for a few days. Since these interruptions of temperature arise from a different distribution of atmospheric pressure from what usually prevails, they are generally either preceded or followed by stormy weather.

647. If, after an unusual prevalence of south-west wind, or the equatorial current, the polar current or north-east wind should set in, it is probable that easterly winds will prevail for some time. If the season be winter, frost, and perhaps snow, may be looked for; and if summer, the weather will become dry, warm, and bracing, particularly if the wind be E. or S.E. But suppose easterly winds have largely predominated in autumn, and south-westerly winds begin to prevail in the end of November or beginning of December, the weather is likely to continue exceptionally mild, with frequent storms of wind and rain, till about Christmas. This period occurs nearly every year, and its beginning is popularly known as St Martin's summer. On the same principle, if easterly winds preponderate largely above the average in spring, the summer is likely to be characterised by southwesterly winds, with much rain and moisture, and little sunshine ; but if easterly winds nearly fail in spring, they are likely to prevail in summer, bringing in their train dry, warm, bracing weather, clear skies, and brilliant sunshine. This latter prognostic is confidently believed in by several meteorologists; but a comparison of the weather of spring with that of the succeeding summer during the last eleven years in Scotland shows, that while it holds good in the majority of cases, the number of times it fails are too many to entitle it to the claim of a trustworthy prognostic.

648. It is an opinion which has been long and popularly entertained that the changes of the moon have so great an influence on the weather that they may be employed with considerable confidence in prediction. That the moon's changes exercise an influence so strongly marked as to make itself almost immediately felt in bringing about fair or rainy, or settled or stormy, weather, an examination of meteorological records, extending over many years, conclusively disproves.

649. Sir John Herschel states, in his . Familiar Lectures,' that the moon has a tendency“ to clear the sky of cloud, and to produce, not only a serene but a calm night, when so near the full as to appear round to the eye." Arago says, “ La lune mange les nuages.” If these opinions are founded on fact, then the moon must have a very great and decided influence in clearing the sky of clouds. William Ellis has examined the Greenwich meteorological records from 1841 to 1847, and shown from these seven years' observations that such a peculiar and striking effect does not exist.* The popular opinion probably arises from the circumstance that the

* • Philosophical Magazine' for July 1868.

clearing of the sky near the time of full moon arrests the attention, whereas the clearing of the sky when the moon is not present is less likely to be noticed.

650. From a most laborious investigation, which meteorologists alone can adequately appreciate, Park Harrison has shown that shortly after full moon there is a tendency to dispersion of cloud, which, though not very marked, is yet appreciable; and he has further shown, from the observations of temperature at Greenwich for 1841-47 and 1856-64, at Oxford for 1856-64, and at Berlin for 1820-35, that a maximum mean temperature occurs on the average at each of these places on the 6th and 7th day of the lunation, when the moon's crust turned towards the earth is coldest ; and a minimum mean temperature shortly after full moon, when the moon's crust having been exposed for some days to the sun's heat is warmest. The conclusion has been drawn that the lowering of the temperature immediately after full moon is caused by the partial clearing of the sky of cloud by the higher temperature caused by the full moon, by means of which terrestrial radiation is less impeded, and the temperature consequently falls. Now, it is necessary to distinguish here between the facts of observation and the conclusions drawn from them. The lower averages of cloud and of temperature after full moon, and the higher averages in the moon's first quarter, are interesting facts in the meteorology of the places for which the averages were taken; and they are also valuable as suggesting further inquiry; but they do not warrant the broad conclusion which has been deduced from them.

651. Schübler has examined sixteen years' observations of the wind, and has found that the S. and W. winds increase in frequency from new moon to the second octant, whilst in the last quarter the same winds are at a minimum, and N. and E. winds reach their maximum. Glaisher has generally confirmed these results, from a discussion of the Greenwich observations of the wind from 1841 to 1847.* Let it be supposed that this relation of the winds to the phases of the

* ‘Proceedings of Meteorological Society of England,' March 1867.

moon were established, it would then be unnecessary to resort to dispersion of cloud and increased terrestrial radiative in order to account for the lower average temperature, seeing that the greater prevalence of northerly and easterly winds would be amply sufficient to bring about these results.

652. If it be the case that there is an immediate connection between the phases of the moon and these changes of temperature, then it necessarily follows that the same relation may be observed anywhere over the world. It is self-evident, especially to those who have charted the weather for considerable portions of the earth's surface, that if Schübler's and Glaisher's conclusions regarding the wind hold good for France and the south of England, the same winds will not prevail at the same time over even so small a portion of the earth's surface as Europe. These winds, it will be observed, possess different qualities, the one being moist and warm, and the other dry and cold; hence they point to a different distribution of the barometric pressure. The proving of the monthly recurrence of such distributions of pressure affecting the winds, and, through them, the temperature at Greenwich; the tracing of these perpetually recurring changes to lunar influence; and the extension of the inquiry into other parts of the world which would be necessary before any general result could be arrived at, present a problem so vast and so laborious that few would care to encounter it. Joseph Baxendell* has examined the St Petersburg observations for the years 1856-64, the years for which Mr Harrison examined the Greenwich and the Oxford temperature. There were 111 lunations during these nine years, and the mean temperatures were found to be as follows:-Day after New Moon. Mean Temp. | Day after Full Moon. Mean Temp. 5th day,

3d day,


7th ,

6th ,


39.96 Mean, 38.20

Mean, 39.31

[ocr errors]
[ocr errors]
[ocr errors]
[ocr errors]

Proceedings of Lit. and Phil. Society of Manchester,' 1867-68.

« AnteriorContinuar »